Hypothesis: In adults starting allopurinol, a pre-treatment model that combines HLA-B*58:01 status, eGFR, starting dose, and thiazide exposure will discriminate severe allopurinol hypersensitivity risk better than genotype-only screening.

Why this matters: catastrophic reactions are rare, but bedside failures usually occur when modest risks stack rather than when one variable acts alone.

Testable prediction: in a prospective multicenter new-user cohort, the composite model will improve 90-day SCAR/AHS discrimination versus HLA-B*58:01 alone by >=0.05 AUC and improve net reclassification for clinically avoidable starts.

Suggested design: prospective or target-trial-emulation cohort of incident allopurinol users with ancestry, genotype, renal function, starting dose, co-medications, and adjudicated 90-day SCAR/AHS outcomes. Compare calibration, Brier score, AUC, decision curves, and net benefit.

Falsification: if genotype-only screening matches or exceeds composite-model discrimination and calibration, the hypothesis fails.

References: FitzGerald JD et al. Arthritis Rheumatol. 2020;72(6):879-895. DOI:10.1002/art.41247. Goncalo M et al. Br J Dermatol. 2013;169(3):660-665. DOI:10.1111/bjd.12389. Lu CY et al. J Formos Med Assoc. 2019;118(1 Pt 1):206-214. DOI:10.1016/j.jfma.2018.06.006.